Thursday, July 25, 2013

LHCb: \(3\)- or \(4\)-\(\sigma\) excess of \(B\)-mesons' muon decays

New physics may already be looking at us

Tommaso Dorigo has discussed something that may be interesting – a hint of new physics coming from the LHCb experiment:
A Four-Sigma Evidence Of New Physics In Rare B Decays Found By LHCb, And Its Interpretation
The deviation will be described in the LHCb-PAPER-2013-037 paper, now in preparation. Locally, it is a \(3.7\)-\(4.0\sigma\) effect which is reduced to \(2.5\)-\(2.8\sigma\) once you take the 24 bins into account (look-elsewhere effect). See page 13 of Nicola Serra's presentation in Stockholm.

Update: A preprint will appear in December 2015, JHEP, ScienceAlert 2016.

We just described how strong the evidence is. But what events is the evidence about? Well, it is about the decay of the neutral \(B\)-mesons\[

B^0 \to K^* \mu^+ \mu^-, \quad K^*\to K^+ + \pi^-.

\] Recall that the asterisk denotes a virtual particle in this experimental jargon. When some "new observables" are used, they see the aforementioned excess of events approximately in bins with the transferred momentum\[

1\GeV\lt \sqrt{|q^2|}\lt 3\GeV

\] or so, especially between \(2\GeV\) and \(3\GeV\).

Dorigo recommends you a three-day-old hep-ph paper by Sebastien Descotes-Genon, Joaquim Matias, and Javier Virto
Understanding the \(B\to K^* \mu^+\mu^-\) Anomaly
This paper is aware of the anomaly we have mentioned and proposes a parameterization of the anomaly – some new physics' contribution to the "Wilson" coefficient \({\mathcal C}_9\) of the semileptonic operator\[

{\mathcal O}_9 = \frac{e^2}{16\pi^2} (\bar s\gamma_\mu P_L b) (\bar \ell \gamma^\mu\ell).

\] The operators with subscripts \(7\) and \(10\) are discussed, too. Who is contributing to the coefficient isn't really clarified so this whole discussion remains somewhat boring and technocratic.

However, some TRF readers with a very good memory should have a deja vu feeling. Haven't we seen something similar?

Yes, we have. First of all, in May 2012, BaBar reported a \(3.4\sigma\) excess in \(\tau\nu\) decays of the B-mesons. But the \(\tau\) leptons were involved so it's a different final state than the final states considered here.

In March 2013, I described some potentially huge, \(7\sigma\) or \(9\sigma\) excesses in pion and kaon decays of the charged \(B\)-mesons seen by the LHCb. Note that journalists often present each experiment that agrees with the Standard Model as strong if not lethal evidence against new physics but they get it upside down: it's enough to find one experiment that disagrees with the Standard Model to falsify that good old theory and it's plausible that we already know such experiments.

Most similarly, in November 2013, Gordon Kane wrote a TRF guest blog about the superstringy predictions for the\[

B^0_s\to \mu^+\mu^-

\] decays which are similar to the decays discussed in this text but they didn't including the virtual kaon. The stringy prediction looked very accurate to them (LHCb and CMS just published their combined data on this very process that are compatible with the Standard Model) and the processes mostly depended on the moduli – the scalar fields encoding the oscillating shape of the compactified dimensions whose particles should be at least as heavy as \(30\TeV\) or so for cosmological or astrophysical reasons.

So some evidence of new physics may already be "out there" but time and independent observations will probably be needed before we will feel any certain about these claims – and, equally if not more importantly, about the type of new physics that causes these anomalies.


  1. Thanks for this nice article Lumo explaining what you think about these results, I was waiting for it :-)

    I dont trust TD (stoped clicking him ...) and as nobody else wrote about this excess first thought he is just trolling us ...

  2. Since I am one of the authors of the technocratic paper.. :) let me clarify a few things. Many people fall in your same mistake (and I can understand it) and Nicola tried to insist on this and we also: we are not talking about a 4 sigma deviation alone (I do not care on one single deviation). You should look into the table of deviations in our boring paper. Our paper is about if there is a pattern of deviations or not and it looks like there is (this is the big difference with all the past deja vu). It is as important TO SEE the deviations in (P2, 2.9, P5prime 2bin 1.6 p5prime 3bin 4 sigma, AFB 2 sigmas) as NOT TO SEE deviations for instance in P4prime. And indeed I bet (I hope) that the 4 sigma will go slightly down... If it happens it will be much more interesting.... Having said that it is time for theorists to look into many different approaches and get different predictions to compare with (that's very important) and also for experimentalist, that did a fantastic job (Nicola is an outstanding experimentalist) to see how the pattern evolves with more statistics. Finally, C9 is the most difficult place from theory point of view but it is also one of the less tested coefficients... In any case exciting times are ahead of it is time to prove or disprove it, ma si muove...

  3. Dear Quim, just to be sure, I think that this "boring" type of work is very important. I also understand it's not just about seeing any excess; it''s important to know what and where the excesses are and so on. Different excesses and/or their combinations have different interpretations.

    What you say about C9 sounds like it's been understudied which is a purely social phenomenon. Model builders should look for C9-related signatures of their models, at least some of them that actually predict these signatures.

  4. Thanks Lubos. I meant that our sensitivity to C9 was low compared to other coefficients. In this sense if new physics could be seen that's the place. Still there is a lot of work to do to make sure (long distances, etc.) and the idea of this work was to pointed out that there is something going on with O9 and that all of the tensions are explained in the same way. This is a first analysis and I expect many more to come. Now many model builders come to me (already at the conference) to discuss on the impact of their model in this coefficient. Still I think that this is not the only surprise hidden in this interesting mode B->K*mumu.